Chemoenzymatic synthesis of a non-peptide tachykinin NK-2 antagonist

Article abstract of DOI:10.1016/S0040-4039(00)01149-7

The synthesis of a tachykinin NK-2 antagonist (S)-11 has been carried out in four steps starting from the the (S)-(+)-enol acetate 3, which was obtained in 100% e.e. by resolution of the racemic ester with Pseudomonas fluorescens lipase. The absolute configuration of the enol acetate (+)-3 was confirmed by X-ray analysis of the camphanyl derivative 13. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01149-7

Tetrahedron Letters 41 (2000) 6929±6933
Chemoenzymatic synthesis of a non-peptide tachykinin
NK-2 antagonist
Andrew J. Carnell,a,* Maria Luisa Escudero Hernandez,a Alan Pettmanb
and Jamie F. Bickleya
aDepartment of Chemistry, Robert Robinson Laboratories, University of Liverpool, Liverpool L69 7ZD, UK
bDepartment of Process Research and Development, P®zer Ltd, Sandwich, Kent CT13 9NJ, UK
Received 9 June 2000; accepted 5 July 2000
Abstract
The synthesis of a tachykinin NK-2 antagonist (S)-11 has been carried out in four steps starting from the
the (S)-(+)-enol acetate 3, which was obtained in 100% e.e. by resolution of the racemic ester with
Pseudomonas ¯uorescens lipase. The absolute con®guration of the enol acetate (+)-3 was con®rmed by X-ray
analysis of the camphanyl derivative 13. # 2000 Elsevier Science Ltd. All rights reserved.
The sensoneuropeptide tachykinins, which include substance P and neurokinins A and B, are
distributed in the peripheral and central nervous systems and are known to be involved in
neurological in¯ammation, pain transmission and bronchoconstriction.1 The e€ects of these
tachykinins are mediated through the activation of NK-1, NK-2 and NK-3 receptors, and new
antagonists are being sought as a means of controlling pain and in¯ammation. P®zer have
developed a class of non-peptidic neurokinin NK-2 antagonists generally consisting of (S)- 5-(3,4-
dichlorophenyl)-5-dialkylaminoethyl-N-alkyl d-lactams as shown in structure 1 (Scheme 1).2
Scheme 1.
* Corresponding author. E-mail: a.j.carnell@liverpool.ac.uk
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01149-7

6930
We envisaged an asymmetric synthesis of this important class of drugs in which the quaternary
chiral centre in the lactam 1 could be accessed from the chiral enol acetate 3. Oxidative cleavage
of the enol double bond in the enol ester would give an intermediate aldehyde ester 2, possessing
the correct (S) absolute con®guration, which upon reductive cyclisation3 of the cyano group onto
the ester would give the required lactam. The aldehyde group could then be reductively aminated
with any secondary amine followed by alkylation of the lactam nitrogen.
We have recently developed the lipase resolution of racemic enol acetates derived from
prochiral ketones where the product ketone can be recycled, leading to greater than 50% yields of
the enol acetate after several cycles (Scheme 2).4,5 For example, the racemic enol acetate 4,
derived from ketone 5, was resolved using Amano Pseudomonas ¯uorescens lipase (PFL) to give
unreacted enantiomerically pure (100% e.e.) (S)-enol acetate 4 and recovered ketone 5.5 We were
pleased to ®nd that biotransformation of the 3,4-dichlorophenyl substrate 36 with the same
enzyme gave a 30% yield of the (+)-enol ester 3 in 100% e.e. after 68% conversion to ketone.
Although the selectivity was not optimal, the resolution could be carried out in batches of up to
10 g under the same conditions as used previously for substrate 4, and both the enzyme and
ketone could easily be recycled.7 In related work, we have found that the active site of PFL can
accommodate changes in the nature of the substitution on the phenyl ring of the substrate
without a€ecting the selectivity signi®cantly.8
Scheme 2. Reagents and conditions: (i) Pseudomonas ¯uorescens lipase, THF, n-BuOH; (ii) isopropenyl acetate, p-
TsOH
As a comparison with the well-established chiral lithium amide methodology, we treated the
ketones 5 and 6 with lithium (R,R)-(^)-bis(a-methylbenzyl)amide at ^78ꢀC, and trapped the
enolate with acetic anhydride to give the enol acetates (R)-4 (54% e.e., 50%) and (R)-3 (57% e.e.,
54%), respectively. Treatment of 4-cyano-4-phenylcyclohexanone with the same base at ^100ꢀC
and trapping with TMSCl gave 64% yield and 64% e.e.
Hence, with these particular substrates, the enzymatic resolution provides the best method for
obtaining enantiomerically pure material.
With a method for the production of the chiral enol ester 3 in hand, we embarked on our
proposed synthesis of a representative member of the NK-2 antagonists, (S)-lactam 11 (Scheme 3).
At this stage, we assumed the (S)-con®guration for (+)-3 by analogy to that previously
established for the enol acetate 4.5 Oxidative cleavage with ozone in methanol:dichloromethane
(1:4) gave the aldehyde-ester (^)-7 in 60% yield. Reductive amination with morpholinoazetidine9
hydrochloride 8 under hydrogenation conditions (Pd±C/H2) was not successful. The presence of
the triethylamine used to release the free azetidine amine from its hydrochloride salt or the
resultant triethylamine hydrochloride may have poisoned the palladium catalyst. However, prior
formation of the imine between 8 and (^)-7 in THF in the presence of triethylamine followed by

6931
Scheme 3. Reagents and conditions: (i) O3, MeOH:CH2Cl2 (1:4), ^78ꢀC then PPh3, rt (60%); (ii) (a) 8, THF, Et3N, rt,
(b) NaB(OAc)3H, AcOH (91%); (iii) H2, raney-Ni, MeOH, 45ꢀC (86%); (iv) NaH, 3-methoxybenzylbromide, DMF,
18-crown-6, 0ꢀC (27%)
reduction with sodium triacetoxyborohydride a€orded the intermediate (^)-9 in 91% overall
yield. The reductive cyclization of (^)-9 to give lactam (+)-10 was achieved in 86% yield using
hydrogenation over raney-nickel. Fortunately, no dechlorination of the phenyl ring was observed
under these conditions. Lactam 10 underwent N-benzylation with 3-methoxybenzyl bromide to
furnish the target NK-2 antagonist (+)-11 in 27% yield. In attempts to improve this step, we used
valerolactam as a model substrate. This underwent N-benzylation cleanly with 3-methox-
ybenzyl bromide or chloride under a variety of conditions, including sodium hydride and 18-
crown-6 in DMF and potassium hydroxide in DMSO/water, giving yields of around 75%.
However, when these conditions were applied to the lactam 10, low yields resulted. The
morpholinoazetidine group may undergo N-benzylation and the resulting alkylammonium species
may not be ecient alkyl donors or may undergo ring cleavage leading to more polar products.
In order to con®rm the suspected (S)-con®guration of the enol acetate (+)-3, we converted the
aldehyde (^)-7 into the crystalline camphanic ester derivative 13 via alcohol 12, as shown in
Scheme 4. Crystallisation of the 0ester 13 from 2-propanol gave crystals suitable for X-ray analysis
(Fig. 1).10 By reference to the 1 (S) camphanyl moiety, the absolute con®guration of the cyano-
substituted centre was determined to be (S), thus con®rming the con®guration of the enol acetate
(+)-3 and the NK-2 antagonist (+)-11 as (S).
In summary, the enantiomerically pure enol acetate (S)-3 was obtained by biotransformation
and used to synthesise the tachykinin NK-2 antagonist (S)-11 in four steps. As previously stated,

6932
Scheme 4. Reagents and conditions: (i) NaB(OAc)3H, AcOH, <20ꢀC (72%); (ii) (^)-1(S)-camphanic acid chloride,
DMAP, Et3N, DCM (83%)
Figure 1. X-Ray crystal structure of camphanyl derivative 13 showing the crystallographic numbering
the ketone product of the biotransformation is prochiral and can easily be separated and chemically
recycled, thus providing greater than 50% yield of the chiral enol ester over several cycles. We are
currently developing an in situ method for this recycling, which will be reported elsewhere.
Acknowledgements
We would like to thank P®zer Ltd for ®nancial support and Amano Enzyme Europe Ltd for
the kind donation of the Pseudomonas ¯uorescens lipase.
References
1. For reviews, see: Maggi, C. A.; Manzini, S. Drugs and the Lung 1994, 507. Rees, D. Annu. Rep. Med. Chem. 1993,
28, 59. Longmore, J.; Swain, C. J.; Hill, R. G. Drug News and Perspectives 1995, 8, 5. Fauchere, J. J. Pharmacol.
Rev. 1994, 46, 551.
2. P®zer Patent, WO 96/05193-A1. CA No. 125:86674.
3. Chen, H. G.; Chung, F.-Z.; Goel, O. P.; Johnson, D.; Keston, S.; Knobelsdorf, J.; Lee, H. T.; Rubin, J. R. Bioorg.
Med. Chem. Lett. 1997, 7, 555.

6933
4. Carnell, A. J.; Swain, S. A.; Bickley, J. F. Tetrahedron Lett. 1999, 40, 8633.
5. Carnell, A. J.; Barkley, J.; Singh, A. Tetrahedron Lett. 1997, 38, 7781.
6. Dei, S.; Novella Romanelli, M.; Scapecchi, S.; Teodori, E.; Chiarini, A.; Gualtieri, F. J. Med. Chem. 1991, 34,
2219.
7. Preparative scale biotransformation: the enol acetate 3 (10 g, 32.4 mmol), P. ¯uorescens lipase (8 g) and n-BuOH
(5.21 ml, 64.7 mmol) were stirred in tetrahydrofuran at room temperature for 9.5 h, whereupon HPLC (Chiracel
AD) indicated 100% e.e. for the enol acetate. Rt: (R)-3 15.5 min; (S)-3 20.9 min; eluent 100% EtOH, 0.5 ml/min.
The solution was ®ltered through a glass sinter funnel, the residual enzyme washed with tetrahydrofuran and the
solvent removed by evaporation under reduced pressure. The crude residue was puri®ed by ¯ash column
chromatography (eluent, diethyl ether:petroleum ether, 1:2) to give the ketone 6 (7 g) and the (S)-enol acetate 3
(2.8 g, 28%) as a white solid (HRMS: found M+ 309.03227. C15H13ClO2N requires 309.03235); mp 153.5ꢀC; ꢀmax
(neat) 2233 (CN), 1755 (CˆO) cm^1; ꢁH (300 MHz, CDCl3): 2.23 (3H, s, COCH3), 2.23±2.67 (6H, m, 2ÂH-3, 2ÂH-
5 and 2ÂH-6), 5.46 (1H, s, H-2), 7.22±7.56 (3H, Ar); ꢁC (75 MHz, CDCl3) 20.9 (CH3), 24.6, 32.7, 35.4 (C-3, C-5
and C-6), 40.0 (C-4), 110.5 (C-2), 121.5 (CN), 125.3, 128.1, 131.1 (C-20, C-50, C-60), 132.8, 133.5, 139.6 (C-10, C-30
and C-40), 148.1 (C-1), 169.3 (CˆO); m/z (EI) 309 (3%, M+), 267 (11), 70 (69), 43 (100).
8. Allan, G.; Carnell, A. J.; Escudero Hernandez, M. L. Manuscript in preparation.
9. Morpholinoazetidine 8 was kindly donated by P®zer Ltd.
10. Sheldrick, G. M. SHELX97, Universitat Gottingen, 1997. Crystal data for C24H27Cl2O6 (13): M=496.42,
orthorhombic, a=7.3163(8), b=10.194(2), c=32.052(4) A, U=2390.6 A3, T=213 K, space group P212121, Z=4,
m(Mo-K)=0.263 mm^1, Stoe IPDS di€ractometer, 15326 re¯ections measured, 3770 unique (Rint=0.1987), which
were used in all calculations. R1 (F>4F)=0.0468 wR(F2), 0.1103 (all data). Single crystals of compound 13 were
obtained by recrystallisation from 2-propanol, mounted in inert oil and transferred to a cold gas stream in the
di€ractometer. The structure was solved using direct methods and re®ned by full least-squares on F2.